• Abstract and Figures
• Public Full-text

CS 6110 S18 Lecture 17 Predicate Transformers 1 Relative Completeness In the last lecture, we discussed the issue of completeness|i.e., whether it is possible to derive every valid partial correctness specification using the axioms and rules of Hoare logic. Unfortunately, if treated as a pure deductive system, Hoare logic cannot be complete. To see why, consider the following partial correctness specifications: ftrueg skip fP g ftrueg c ffalseg The first is valid if and only if the assertion P is valid while the second is valid if and only if the command c does not halt. It turns out that the culprit is the weakening rule, which includes premises involving the validity of implica- tions between the assertions involved: ' ) '0 f'0gcf 0g 0 ) (weakening) . f'gcf g Although we cannot have a complete proof system for first-order formulas, Hoare logic does enjoy the property stated in the following theorem: Theorem 1. 8'; ; c: f'g c f g implies ` f'g c f g: This result, due to Cook (1974), is known as the relative completeness of Hoare logic. It says that Hoare logic is no more incomplete than the language of assertions|i.e., if we had an oracle that could decide the validity of assertions, then we could decide the validity of partial correctness specifications. 2 Weakest Preconditions Given a program s and a postcondition ', the weakest property of the input state that guarantees that s halts in a state satisfying ', if it exists, is called the weakest precondition of s and ' and is denoted wp s '. This says that • wp s ' implies that s terminates in a state satisfying ' (wp s ' is a precondition of s and '), • if is any other condition that implies that s terminates in a state satisfying ', then ) wp s ' (wp s ' is the weakest precondition of s and '). As in the λ-calculus, juxtaposition represents function application, so wp can be viewed as a higher-order function that takes a program s and a postcondition ' and returns the weakest precondition of s and '. The function wp can also be viewed as taking a program and returning a function that maps postconditions to preconditions. For this reason, axiomatic semantics is sometimes known as predicate transformer semantics. 3 Weakest Liberal Preconditions Cook's proof of relative completeness depends on the notion of weakest liberal preconditions. Given a com- mand c and a postcondition the weakest liberal precondition is the weakest assertion ' such that f'g c f g 1 is a valid triple. Here, \weakest" means that any other valid precondition implies '. That is, ' most accurately describes input states for which c either does not terminate or ends up in a state satisfying . Formally, an assertion ' is a weakest liberal precondition of c and if: 8σ; I: σ I ' () (C c σ) undefined _ (C c σ) I J K J K We write wlp(c; ) for the weakest liberal precondition of command c and postcondition . From left-to-right, the formula above states that wlp(c; ) is a valid precondition: f'g c f g. The right-to-left implication says it is the weakest valid precondition: if another assertion R satisfies fRg c f g, then R implies '. It can be shown that weakest liberal preconditions are unique modulo equivalence. We can calculate the weakest liberal precondition of a command as follows: wlp(skip;') = ' wlp((x := a; ') = '[a=x] wlp((c1; c2);') = wlp(c1; wlp(c2;')) wlp(if b then c1 else c2;') = (b =) wlp(c1;')) ^ (:b =) wlp(c2;')) The definition of wlp(whilebdoc; ') is slightly more complicated|it encodes the weakest liberal precondition for each iteration of the loop. To give the intuition, first define the weakest liberal precondition for a loop that termintes in i steps as follows: F0(') = true Fi+1(') = (:b =) ') ^ (b =) wlp(c; Fi('))) We can then express the weakest liberal precondition using an infinitary conjunction: ^ wlp(while b do c; ') = Fi(') i See Winskel Chapter 7 for the details of how to encode the weakest liberal precondition for a while loop as an ordinary assertion. To check that our definition is correct, we can prove (how?) that it yields a valid partial correctness specification: Lemma 2. 8c; : fwlp(c; )g c f g and 8ρ. fρg c f g implies (ρ =) wlp(c; )) It is not hard to prove that it also yields a provable specification: Lemma 3. 8c; : ` fwlp(c; )g c f g Relative completeness follows by a simple argument: Proof Sketch. Let c be a command and let P and be assertions such that the partial correctness specification fP g c f g is valid. By Lemma 2 we have P =) wlp(c; ). By Lemma 3 we have ` fwlp(c; )g c f g. We conclude ` fP g c f g using weakening. 2.

Load more

  2

Copy Link